I think the way to approach bargains is the to look at projected prices at the league tendency and then compare that price to each player based on the actual scoring split. By doing so, you essentially get a projected ROI (positive and negative) for each player, correct? What I'm sure you'll find is that at the high level, the ROI substantially negative compared to their actual value, and vice versa as you move down.
If this were me doing this, I would pick the level at which I can "save" money by buying some players below their ROI.
I would then take my savings and apply it to higher level guys, even if that means overpaying based on ROI.
Let me see if I can come up with an example (totally making up numbers here)...
OF - Lg$ - Proj$
Sizemore - $40 - $32
Upton - $35 - $30
Lee - $28 - $28
Ibanez - $22 - $24
Abreu - $16 - $20
Dukes - $10 - $15
So your analysis reveals that you should be able to get a guy like Abreu, Ibanez, and/or Dukes at ROIs of 9% ($2), 25% ($4), and 50% ($5) respectively. You, me, and everyone else know that you're limited in starter spots, so just loading up on cheap players won't win you a championship. You have to grab some big guns at some point, which means spending. Looking at the top of the chart, you see that a guy like Upton will go for an ROI of -14% (-$5). If you acquired Abreu, Lee, & Upton, your combined ROI would be -1% ($1). Then you have to ask yourself: Is that team better than a top guy, like Sizemore, and some lower-level guys.
In the end, what you're doing is not so much purchasing based solely on price. Instead, you're determining where your savings will come later so as to give you a leg up early in your bidding. By projecting how much increased savings you'll have later in the auction, you'll be able to know how much of a loss you can take early on without hurting your team. Hopefully, you end up doing a better job than someone else who's simply doing that on the fly.